Leon/docutranslate
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix

Browse Source
This commit is contained in:
xunbu
2025-08-24 12:02:56 +08:00
parent 486ac3c90a
commit 5cd0156978
6 changed files with 51 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
docutranslate/app.py
View File

@@ -1,7 +1,6 @@
import asyncio
import base64
import binascii
import io
import logging
import os
import shutil
@@ -12,13 +11,12 @@ import uuid
from contextlib import asynccontextmanager, closing
from pathlib import Path
from typing import List, Dict, Any, Optional, Literal, Union, Annotated, TYPE_CHECKING, Type
from urllib.parse import quote
import httpx
import uvicorn
from fastapi import FastAPI, HTTPException, APIRouter, Body, Path as FastApiPath
from fastapi.openapi.docs import get_swagger_ui_html, get_swagger_ui_oauth2_redirect_html, get_redoc_html
from fastapi.responses import HTMLResponse, JSONResponse, StreamingResponse, FileResponse
from fastapi.responses import HTMLResponse, JSONResponse, FileResponse
from fastapi.staticfiles import StaticFiles
from pydantic import BaseModel, Field, field_validator
@@ -247,7 +245,8 @@ class MarkdownWorkflowParams(BaseWorkflowParams):
mineru_token: Optional[str] = Field(None, description="当 `convert_engine` 为 'mineru' 时必填的API令牌。")
formula_ocr: bool = Field(True, description="是否对公式进行OCR识别。对 `mineru` 和 `docling` 均有效。")
code_ocr: bool = Field(True, description="是否对代码块进行OCR识别。仅 `docling` 引擎有效。")
model_version: Literal["pipeline", "vlm"] = Field("vlm", description="Mineru模型的版本，'vlm'是更新的版本。仅 `mineru` 引擎有效。")
model_version: Literal["pipeline", "vlm"] = Field("vlm",
description="Mineru模型的版本，'vlm'是更新的版本。仅 `mineru` 引擎有效。")
@field_validator('mineru_token')
def check_mineru_token(cls, v, values):
@@ -348,7 +347,8 @@ TranslatePayload = Annotated[
# 4. 创建最终的请求体模型
class TranslateServiceRequest(BaseModel):
file_name: str = Field(..., description="上传的原始文件名，含扩展名。",
examples=["my_paper.pdf", "chapter1.txt", "data.xlsx", "video.srt", "my_book.epub", "index.html"])
examples=["my_paper.pdf", "chapter1.txt", "data.xlsx", "video.srt", "my_book.epub",
"index.html"])
file_content: str = Field(..., description="Base64编码的文件内容。", examples=["JVBERi0xLjQK..."])
payload: TranslatePayload = Field(..., description="包含工作流类型和相应参数的载荷。")
@@ -654,7 +654,7 @@ async def _perform_translation(
await workflow.translate_async()
# 4. 任务成功，生成所有可下载文件并存储
task_logger.info("翻译完成，正在生成结果文件...")
task_logger.info("翻译完成，正在生成临时结果文件...")
temp_dir = tempfile.mkdtemp(prefix=f"docutranslate_{task_id}_")
task_state["temp_dir"] = temp_dir
downloadable_files = {}
@@ -689,7 +689,8 @@ async def _perform_translation(
html_config = Srt2HTMLExporterConfig(cdn=is_cdn_available)
elif isinstance(workflow, EpubWorkflow):
html_config = Epub2HTMLExporterConfig(cdn=is_cdn_available)
export_map['html'] = (lambda: workflow.export_to_html(html_config), f"{filename_stem}_translated.html", True)
export_map['html'] = (lambda: workflow.export_to_html(html_config), f"{filename_stem}_translated.html",
True)
if isinstance(workflow, MDFormatsExportable):
export_map['markdown'] = (workflow.export_to_markdown, f"{filename_stem}_translated.md", True)
export_map['markdown_zip'] = (workflow.export_to_markdown_zip, f"{filename_stem}_translated.zip", False)
@@ -711,14 +712,13 @@ async def _perform_translation(
# 循环生成文件
for file_type, (export_func, filename, is_string_output) in export_map.items():
try:
task_logger.info(f"正在生成 {file_type} 文件: {filename}")
content = await asyncio.to_thread(export_func)
content_bytes = content.encode('utf-8') if is_string_output else content
file_path = os.path.join(temp_dir, filename)
with open(file_path, "wb") as f:
f.write(content_bytes)
downloadable_files[file_type] = {"path": file_path, "filename": filename}
task_logger.info(f"成功生成 {file_type} 文件于: {file_path}")
task_logger.info(f"成功生成 {file_type} 文件")
except Exception as export_error:
task_logger.error(f"生成 {file_type} 文件时出错: {export_error}", exc_info=True)
@@ -760,7 +760,7 @@ async def _perform_translation(
if task_state["error_flag"] and temp_dir and os.path.isdir(temp_dir):
shutil.rmtree(temp_dir)
task_logger.info(f"因任务失败，已清理临时目录: {temp_dir}")
task_logger.info(f"因任务失败，已清理临时目录")
task_state["temp_dir"] = None
task_logger.info(f"后台翻译任务 '{original_filename}' 处理结束。")
@@ -1199,7 +1199,8 @@ async def service_content(
file_info = task_state.get("downloadable_files", {}).get(file_type)
if not file_info or not os.path.exists(file_info.get("path")):
raise HTTPException(status_code=404, detail=f"任务 '{task_id}' 不支持获取 '{file_type}' 类型的内容，或文件已丢失。")
raise HTTPException(status_code=404,
detail=f"任务 '{task_id}' 不支持获取 '{file_type}' 类型的内容，或文件已丢失。")
file_path = file_info["path"]
filename = file_info["filename"]
@@ -1299,7 +1300,8 @@ async def temp_translate(
decoded_content = file_content.encode('utf-8')
try:
workflow_config = MarkdownBasedWorkflowConfig(
convert_engine="mineru", converter_config=ConverterMineruConfig(mineru_token=mineru_token, model_version=model_version),
convert_engine="mineru",
converter_config=ConverterMineruConfig(mineru_token=mineru_token, model_version=model_version),
translator_config=MDTranslatorConfig(base_url=base_url, api_key=api_key, model_id=model_id,
to_lang=to_lang, custom_prompt=custom_prompt, temperature=temperature,
thinking=thinking, chunk_size=chunk_size, concurrent=concurrent),

5
docutranslate/converter/converter_identity.py
View File

@@ -1,4 +1,7 @@
from docutranslate.converter.base import Converter
from dataclasses import dataclass
from typing import Hashable
from docutranslate.converter.base import Converter, ConverterConfig
from docutranslate.ir.document import Document

13
docutranslate/converter/x2xlsx/converter_csv2xlsx.py
View File

@@ -1,17 +1,24 @@
import asyncio
import csv
import logging
from dataclasses import dataclass
from io import BytesIO, StringIO
from typing import Hashable
# 引入 chardet 用于编码检测
import chardet
import openpyxl
from docutranslate.converter.x2xlsx.base import X2XlsxConverter
from docutranslate.converter.x2xlsx.base import X2XlsxConverter, X2XlsxConverterConfig
from docutranslate.ir.document import Document
# 配置一个基本的日志记录器（如果您的项目尚未配置）
# logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
@dataclass(kw_only=True)
class ConverterCsv2XlsxConfig(X2XlsxConverterConfig):
def gethash(self) -> Hashable:
return "1"
class ConverterCsv2Xlsx(X2XlsxConverter):
@@ -25,6 +32,8 @@ class ConverterCsv2Xlsx(X2XlsxConverter):
- 完善的错误处理和日志记录。
"""
def __init__(self, config: ConverterCsv2XlsxConfig):
super().__init__(config=config)
def convert(self, document: Document) -> Document:
"""

2
docutranslate/translator/ai_translator/xlsx_translator.py
View File

@@ -1,5 +1,5 @@
import asyncio
from dataclasses import dataclass, field
from dataclasses import dataclass
from io import BytesIO
from typing import Self, Literal, List, Optional

28
docutranslate/workflow/xlsx_workflow.py
View File

@@ -1,11 +1,12 @@
import asyncio
from dataclasses import dataclass
from pathlib import Path
from typing import Self, Type
from typing import Self
from docutranslate.converter.base import ConverterConfig
from docutranslate.converter.converter_identity import ConverterIdentity
from docutranslate.converter.x2xlsx.base import X2XlsxConverter
from docutranslate.converter.x2xlsx.converter_csv2xlsx import ConverterCsv2Xlsx
from docutranslate.converter.x2xlsx.converter_csv2xlsx import ConverterCsv2Xlsx, ConverterCsv2XlsxConfig
from docutranslate.exporter.base import ExporterConfig
from docutranslate.exporter.xlsx.xlsx2csv_exporter import Xlsx2CsvExporter
from docutranslate.exporter.xlsx.xlsx2html_exporter import Xlsx2HTMLExporterConfig, Xlsx2HTMLExporter
@@ -24,11 +25,6 @@ class XlsxWorkflowConfig(WorkflowConfig):
class XlsxWorkflow(Workflow[XlsxWorkflowConfig, Document, Document], HTMLExportable[Xlsx2HTMLExporterConfig],
XlsxExportable[ExporterConfig], CsvExportable[ExporterConfig]):
_converter_factory: dict[
str, Type[X2XlsxConverter | ConverterIdentity]] = {
".csv": ConverterCsv2Xlsx,
".xlsx": ConverterIdentity
}
def __init__(self, config: XlsxWorkflowConfig):
super().__init__(config=config)
@@ -36,17 +32,25 @@ class XlsxWorkflow(Workflow[XlsxWorkflowConfig, Document, Document], HTMLExporta
for sub_config in [self.config.translator_config]:
if sub_config:
sub_config.logger = config.logger
self._converter_factory: dict[
str, tuple[
type[X2XlsxConverter | ConverterIdentity], ConverterConfig|None]] = {
".csv": (ConverterCsv2Xlsx, ConverterCsv2XlsxConfig(logger=self.logger)),
".xlsx": (ConverterIdentity,None)
}
def _get_document_xlsx(self, document: Document) -> Document:
suffix = document.suffix
converter_type = self._converter_factory.get(suffix)
if converter_type is None:
converter_types = self._converter_factory.get(suffix)
if converter_types is None:
raise ValueError(f"Xlsx工作流不支持{suffix}格式文件")
converter = converter_type()
converter_type, converter_config = converter_types
converter = converter_type(converter_config)
return converter.convert(document)
def _pre_translate(self, document_pre_transalte: Document):
document = document_pre_transalte.copy()
def _pre_translate(self, document_pre_translate: Document):
document = document_pre_translate.copy()
translate_config = self.config.translator_config
translator = XlsxTranslator(translate_config)
return document, translator